arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1904.01033cs.LGstat.ML
收藏

多任务软选项学习

Multitask Soft Option Learning

University of Oxford(牛津大学);Delft University of Technology(代尔夫特理工大学)

Maximilian Igl, Andrew Gambardella, Jinke He, Nantas Nardelli, N. Siddharth, Wendelin Böhmer, Shimon Whiteson

摘要:

我们提出了多任务软选项学习(MSOL),一个分层的多任务框架的基础上规划推理。MSOL扩展了选项的概念,为每个任务使用单独的变分后验,通过共享先验进行正则化。这种“软"版本的选项避免了多任务设置中训练过程中的几种不稳定性,并提供了一种自然的方法来学习选项内策略及其终止。此外,它允许微调新任务的选项,而不会忘记它们的学习策略,从而在不降低分层策略的表现力的情况下实现更快的训练。我们的经验表明,MSOL显着优于分层和平面迁移学习基线。

英文摘要:

We present Multitask Soft Option Learning(MSOL), a hierarchical multitask framework based on Planning as Inference. MSOL extends the concept of options, using separate variational posteriors for each task, regularized by a shared prior. This ''soft'' version of options avoids several instabilities during training in a multitask setting, and provides a natural way to learn both intra-option policies and their terminations. Furthermore, it allows fine-tuning of options for new tasks without forgetting their learned policies, leading to faster training without reducing the expressiveness of the hierarchical policy. We demonstrate empirically that MSOL significantly outperforms both hierarchical and flat transfer-learning baselines.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑